If AI Is Creating So Many Jobs, Why Is Nvidia’s CEO Telling People to Become Plumbers?

When Jensen Huang, the CEO of Nvidia, says it’s a “great time to be a plumber,” it initially sounds like a reassuring throwaway line — the kind of comment meant to calm nerves in an anxious moment.

But Huang isn’t just any executive. He runs the company at the center of the global AI boom. His chips power the systems reshaping how work gets done across industries. So when he singles out plumbers as winners of the AI era, the comment lands differently — especially for people whose jobs exist mostly behind screens.

Huang made the remark this week at the World Economic Forum in Davos, as business leaders once again debated whether artificial intelligence will ultimately create more jobs than it destroys. Huang’s answer was optimistic. But the kinds of jobs he chose to highlight were telling.

Rather than software engineers or data scientists, he pointed to plumbers, electricians, construction workers, and steelworkers — the people needed to physically build the data centers and chip factories that AI depends on.

“It’s wonderful that the jobs are related to tradecraft,” Huang said, describing what he called the largest infrastructure buildout in human history.

For people already working in the trades, it sounded like validation. For many white-collar workers watching AI creep into their industries, it sounded like something else entirely.

A Boom — But Not the One People Expected

Huang said demand for manual labor tied to AI infrastructure is surging in the US, with wages in some cases nearly doubling. In certain regions, workers building chip plants and data centers are earning six-figure salaries.

That’s real money. And it’s real opportunity.

But it also quietly flips a long-standing assumption on its head. For decades, workers were told that education, specialization, and digital skills were the safest way to future-proof a career. Now, one of the most powerful figures in tech is suggesting that the most secure jobs in the AI economy may be the ones least touched by software.

If that’s true, it raises an uncomfortable question: what happens to everyone who followed the old advice?

The marketing manager.
The HR director.
The analyst who did everything right and still isn’t sure where they fit in an AI-accelerated workplace.

The Anxiety Beneath the Optimism

Huang has consistently argued that AI isn’t a mass job killer — that it changes work rather than eliminates it. At Davos, he again pointed to radiology as an example: AI has automated certain tasks, but employment in the field has grown.

That argument offers some reassurance. But it also comes with a quieter reality that’s harder to ignore.

Jobs don’t have to disappear to become unstable. They can shrink, fragment, lose status, or demand constant reskilling just to stay viable. Many workers are already living that version of disruption — still employed, but less certain, less secure, and more replaceable than before.

Against that backdrop, Huang’s praise for hands-on trades feels less like a casual observation and more like a signal about where disruption may hit hardest.

A Future That Isn’t Easy to Pivot Into

There’s another tension buried in the “become a plumber” message.

Trade jobs are real, valuable, and increasingly well paid — but they’re also physically demanding, location-specific, and not easily accessible to everyone. A mid-career office worker can’t always pivot into construction work, no matter how strong the demand or how high the wages.

Reskilling sounds simple in theory. In practice, it often means starting over — physically, financially, and professionally — at a stage of life when many people can’t afford to.

So while Huang’s vision highlights opportunity, it also quietly shifts the risk of transition onto workers themselves.

Davos Confidence, Everyday Uncertainty

Huang’s comments were delivered on a Davos stage alongside BlackRock CEO Larry Fink, in a setting defined by confidence, optimism, and big-picture thinking. Outside that environment, the mood is less certain.

Tech layoffs continue. White-collar hiring has cooled. Workers across industries are being told to adapt faster, learn more, and prove their value in ways that didn’t exist a few years ago.

In that context, the idea that plumbers are among the safest bets in the AI economy feels both empowering and unsettling at the same time.

The Question That Doesn’t Go Away

Huang insists that “everybody should be able to make a great living” in the AI era — and he may be right. But if the most practical advice from one of AI’s most influential architects is to work with your hands rather than your head, it leaves a question hanging in the air.

If AI is creating so much opportunity, why does the path to security look so unfamiliar — and so far removed from the careers so many people were told to build?

That’s the tension at the heart of the AI boom. And it’s one that optimism alone doesn’t quite resolve.

Top 10 Jobs AI Is Unlikely to Replace Anytime Soon

As artificial intelligence spreads across offices, screens, and workflows, one reality is becoming clearer: the jobs hardest for AI to replace aren’t always the most technical ones. They’re the roles that rely on physical presence, human judgment, trust, or emotional nuance — things algorithms still struggle to replicate.

Here are 10 types of jobs that remain relatively insulated from full AI replacement, at least for now:

Plumbers and Electricians
Physical problem-solving in unpredictable environments remains one of AI’s biggest weaknesses. Every building is different, every issue is situational, and mistakes are costly.
Construction and Skilled Trades Workers
From welding to steelwork, these roles require on-site judgment, dexterity, and coordination that robotics hasn’t mastered at scale.
Nurses and Care Workers
AI can assist with diagnostics and paperwork, but caregiving depends on empathy, trust, and real-time human response — especially in high-stress situations.
Teachers (Especially Early Education)
Instruction isn’t just information delivery. It’s emotional awareness, adaptability, and human connection — areas where AI still falls short.
Therapists and Mental Health Professionals
While AI can simulate conversation, it cannot replace the human rapport and ethical responsibility central to mental healthcare.
Emergency Responders
Firefighters, paramedics, and disaster-response teams operate in chaotic, fast-changing conditions that automation struggles to handle safely.
Skilled Mechanics and Technicians
Diagnosing physical systems often involves intuition built through experience — something AI can assist with but not fully replicate.
Managers and People Leaders
AI can analyze performance, but leading humans — resolving conflict, motivating teams, making judgment calls — remains deeply human work.
Creative Directors and High-Level Creatives
AI can generate content, but setting vision, taste, and cultural context still depends on human sensibility.
Entrepreneurs and Small Business Owners
Building something from nothing requires risk-taking, improvisation, and personal judgment — traits AI can support but not replace.

Why This List Matters Now

What’s striking about this list is how closely it overlaps with the jobs Nvidia’s CEO highlighted in Davos. As AI accelerates, career safety may hinge less on how digital your job is — and more on how human it remains.

That doesn’t mean these roles are immune to change. Many will use AI tools. Some will evolve. But for now, they sit on the far side of the automation line — the same line Huang implicitly pointed to when he said it’s a good time to be a plumber.

And for workers trying to make sense of where they fit in an AI-driven future, that distinction suddenly matters a lot.